Manifold steering for neural network control

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• What is the right geometry for enabling principled steering of neural network behavior?question0.799
  The reframed steering problem the paper introduces
• Steering along manifolds provides better control than linear steering when the concept geometry is non-linear.claim0.797
  The central thesis of the paper, motivating the shift from linear to geometry-aware manifold steering.
• Manifold-respecting steering produces smooth natural behavioral trajectories while linear steering teleports between non-adjacent concepts.claim0.792
• Manifold-aware steering is non-trivial IP requiring geometric analysis, not a system-prompt implementation.claim0.790
• Cross-task and cross-modal validation of manifold steeringfinding0.785
  The paper demonstrates the bidirectional geometry-behavior relationship across multiple tasks and modalities (language models and video world models)
• Manifold steering demonstrates bidirectional geometry-behavior link in a video world model on tasks with geometry corresponding to physical dynamicsfinding0.783
  Extension of manifold steering validation to video world models and physical dynamics tasks, demonstrating cross-modal generality
• Manifold Steering (Wurgaft)method0.758
  Internal-state feedback technique for steering language models; same conceptual mechanism applied by Hazra et al. to chemistry.
• Networks compute on geometric manifolds and control should respect that geometry.claim0.757
  Strong interpretive assertion linking discovery and control: neural computation is fundamentally manifold-structured.